As computer systems and networked computer systems proliferate, information access time becomes more critical. For many reasons, access time to information databases has a tendency to increase even with frequent equipment upgrades and technology advances. This tendency is apparent in mostly all database accesses which involve extensive networks and mirrored networks which may contain the same information in a plurality of separate storage devices and files.
In many applications, desired information files are maintained in tact at several data storage devices such as, but not limited to hard drive storage units. In such cases, each copy is a so-called "data mirror", i.e. one of several listed devices which contain copies of the same information file which is sought by the user. Each device may be served by the same server or a different server in the same room or a different server in another city or even in another country. The access times to any of the mirror devices would depend on a number of factors including the read-out speed of each storage device.
A plurality of mirrored devices, such as disk drives, may also exist within a single system. In many applications, for both network and single system configurations, there are a plurality of disk drives and many of the disk drives contain many of the same files. On occasion, when a particular device is accessed, or if a system arbitrarily selects one device of a plurality of mirror devices to fill a request for a selected file, the device selected may not be the fastest device of those containing the requested file, or, the device may be equally fast, but the data busses may be of non-equal bandwidth. Moreover, if the non-accessed devices are not being used, those information resources are not being optimally utilized.
Accordingly, there is a need for an enhanced method and processing apparatus which is effective to determine an optimized methodology for accessing and returning a desired data file in a minimal period of time from a plurality of devices containing the desired data file.